RedBeard's Curious Life

I am hopelessly intrigued by BitCoin. I made a previous post about it where I spelled out a number of benefits and drawbacks. Current volatile upward velocity (yes, 75% price increase in 2 days) suggests a strong speculative interest, although it could be an attempt to corner the market. Let’s take a look at a few scenarios and what impact they might have on a potential investment into the BitCoin “economy”.

0) Insufficient adoption as currency. Early adopters cash-out, bubble bursts, speculators lose out, everybody forgets about BTC. The biggest hurdle here is the natural deflationary nature of BTC, which motivates people to hold rather than spend them. Merchants are strongly motivated to accept a deflationary currency because their goods will retain their current value but the currency will be worth more tomorrow, but merchants won’t get any BTC if nobody is willing to spend them.

1) Widespread adoption: 10% of world cash currency transactions replaced with BitCoin, 1% of world wealth stored as BitCoin holdings. Target price (current USD): $1000-$1,000,000 per BTC. Obviously this is what speculators are hoping for, but it’s quite a gamble what the target is and when it might happen (if at all).

2) Black market adoption for cash-only workers, drugs, weapons, slave trade, and other questionable activities. 10% of black-market transactions conducted via BitCoin. Target price: about $1000 per BTC. Obviously the purveyors of BitCoin would rather not be associated strictly with organized crime, that sort of thing gets you fined, locked up, or worse (see e-gold for an example).

3) Government buys up a large amount of the BitCoin money supply like gold has been (Ft Knox). Rampant deflation and market instability, currency is no longer immune to fractional reserve banking because it’s effectively controlled by a government. The act of buying up enough BTC off the market to gain a controlling stake would drive the price up and cripple supply for transactions, but it would also enrich with USD the current holders of BTC; maybe 25% of the market would be enough, so about 5 million BTC which could be bought at current prices for only $100M which is chump change to the US government, but obviously the cost increases as the market shrinks and it could cost a few billion USD to corner the market. The government could take other actions to shut down trading and production of BTC, but that boat may have already sailed.

4) Cryptographic attacks on “work” algorithm. Effectively counterfeit BitCoins could be produced by clients doing an unbalanced amount of work, which would cause inflation for non-counterfeiters and market instability when they are introduced to the market. If the attack were publicized or widely adopted, it would be integrated into the work expectation and the network would re-normalize.

5) Social or cryptographic attacks on transactions or holdings (transaction signing, wallet store, man-in-middle, etc). Market instability, loss of trust in currency, hyperinflation. There is no central or accountable authority for BTC. If someone steals or peeks into your BTC wallet, they are free to spend its entire contents in an irreversible and potentially untraceable way, perhaps just sending it all to themselves at a newly-generated address (or multiple new addresses) which could then be laundered and anonymized. Worse results would occur if the transaction signing mechanism were compromised and collisions could be generated.

6) A competing e-currency gains more market share than BitCoin. This could be as simple as a shadow BTC2 network with tweaked rules like a higher asymptotic production limit, faster/slower production peak, etc. It may be non-deflationary or offer other characteristics which give it greater intrinsic value as a store of wealth or better qualities as a currency of exchange.

I’ve also thought of some more detailed problems since my previous post:

• Mining productivity has a discontinuous rate over time, and this will test the market for block mining when the first reward decrease occurs (from 50 BTC per block to 25, anticipated sometime in 2012 when half of the BTC asymptote has been reached).
• Anonymity is not as simple as advertised. All transactions are stored in the global history and can be traced from one address to another. Obscuring transfers and ownership requires paying fees to move money around, and source transfers from other currency can potentially be traced by subpoena.
• Mining work is only internally valuable, unlike real metal coins, commodities, or services like password hash cracking.
• Wallet is totally unsecured on a computer without specific action by the user (and maybe even then if a debugger is attached).
• Having a single copy of your wallet compromised can mean all your money disappears without a trace.
• Transaction verification time (at least 10 minutes) is too long for reliable practical use. If you don’t wait for verification, the “double-spend” problem is easily exploited.
• Lack of physical medium is a blow against total adoption.
• Loans are difficult to imagine in a deflationary currency, and difficult to exercise (or abuse) in a fixed-reserve currency system. Borrowing 1 BTC @ $100USD and paying back 1.1 BTC @ $200USD is not appealing to a borrower, and a negative interest rate is unappealing to a lender. Perhaps loans will simply be unappealing until the price volatility is reduced.
• Usability of a hash address is pretty terrible. Bank account ACH transfers are about 20 numerical digits, PayPal uses email-addresses, and money orders are sent to physical house addresses. All of those are far easier for a human to remember and use.

For the record, I’m currently shy to the trading market due to extreme volatility and cumbersome barriers to entry (for converting other currency into or out of BTC), but I’m investing a little bit in a mining pool. A few dollars on my electric bill will hardly go noticed if the bottom drops out, but if I can hold onto 10+ BTC and the price continues to rise then I’ll be feeling pretty happy.

Resources:

http://bitcoinweekly.com/articles/bitcoin-what-s-it-backed-by

http://falkvinge.net/2011/06/04/bitcoins-four-hurdles-part-one-usability/

http://www.reddit.com/r/bitcoin

I hadn’t heard of BitCoins a couple of months ago, but they’re currently hitting a few mainstream media outlets. Nobody seems to have a solid handle on them or where they’re headed, but speculation has driven the trading value of BitCoin shares through the roof in the past year. I can’t quite tell if it’s a well-marketed ponzi scheme or a legitimate new-world currency, but I’m leaning quite heavily towards the former.

How do they work?

• Visit bitcoin.org for the semi-official description. Here’s my take…
• BitCoins are produced at a fixed (predictable and dwindling) rate over time by computing cryptographic hashes of various “difficulty”. One “successful” hash block is expected every 10 minutes.
• Transactions of BitCoins are conducted by signing some data with public and private keys, where the public key is used to announce the new owner, and the private key is used to renounce existing ownership. https://en.bitcoin.it/wiki/Transactions
• A successful hash block stores all of the transactions which have transpired in the intervening 10 minutes. https://en.bitcoin.it/wiki/Blocks
• Transactions and successful hash blocks are shared over a peer-to-peer network so they can be “verified” and the next block must refer back to a previous successful hash block, creating a “chain” of transaction histories.
• All machines on the network are racing to complete a successful hash, because they get the 50 BitCoins produced by it. GPUs are faster at hashing than CPUs, but a lone machine is quite unlikely to win the hash lottery; pools of shared machines have cropped up, where many machines cooperate to find a successful hash and then share the output proportionally among the pool members. http://bitcoinx.com/profit/
• The cost of producing a successful hash changes over time according to the hash rate of the network as a whole. A single machine may produce the first million bitcoins over the course of a year, but the next million may consume the resources of a million machines for a whole year, and the next million could be produced by a single machine again (although that last possibility is highly unlikely).

Beneficial Aspects

• Low or no transaction fees
  • Relative to credit cards, which charge at least 3% of the value of the transaction
  • Hash-computing machines may charge transaction fees on all or unusual transactions, of an arbitrary amount
• Potentially anonymous transactions
  • Transactions can be scattered across sender and recipient addresses which make it difficult to trace the owners of the sending and receiving addresses
  • Supposedly useful for buying goods and services on the internet (although the seller still needs to know some details in order to deliver)
  • Perfect for black market trading
• Predictable money supply
  • Supposedly immune to the whims of a central banking authority
  • Implies deflation (increasing value of currency) as the economy expands
• Effective currency by most measures
  • Durable (assuming data is immortal)
  • Portable, high value density
  • Easily divisible
  • Difficult to counterfeit (assuming cryptographic security)
  • Liquid (perhaps)

Problematic Aspects

• Cryptographically unproven
  • The block hashing algorithm searches for a hash in a given range
  • SHA-256 is not known to have specific collision attacks, but a range collision attack is probably far easier
• Too complex for many people to understand, in theory and practice
  • Theory: much more complicated than cash or commodities, and this is before any loans enter the picture
  • Practice: wallet management, transaction security, anonymizing, etc
• BitCoin production (mining) cost increases linearly with demand for production
  • The hash work expectation means that as more individuals enter the BitCoin production market, the higher the per-unit production cost for everybody in the market.
  • In order to sustain production, the value per unit must increase.
  • If the value does not increase sufficiently to provide profit incentive, producers will abandon the market.
  • This scheme heavily favors early adopters who enjoy cheap production and can then liquidate their assets for extreme profit.
• BitCoins have no intrinsic value, but a high production cost
  • Gold is costly to produce (mine) but also has aesthetic and utility value
  • Paper dollars have low production cost and low utility value (its value is derived from government backing)
  • With no intrinsic value, there is no price floor in an exchange, which may invite hyperinflation
• Irreversible currency destruction is possible
  • If you lose your wallet, the money in it is destroyed forever
  • Paper dollars can be printed to shore up money supply
  • Gold can be rediscovered and refined
• No fraud protection
  • Isn’t that what all those credit card fees cover?
  • Cash has similar flaws, but you typically exchange cash and real goods simultaneously in a real-world transaction with the option to back out if the goods are not correct.
• Anonymity is difficult to maintain
  • All transactions are visible to the world
  • If someone knows your public address, they can trace every transaction involving it
• Whole number of bitcoins is too small for practical use
  • The upper limit on the number of whole BitCoins (after 130 years of production) is 21 million
  • microbitcoins are not a friendly unit for people to deal with
• Transaction verification time is too slow for instantaneous purchases
  • One successful hash block every 10 minutes
  • That means AT LEAST a 10 minute wait before that block is distributed and verified
  • That block might not even include your transaction, if the computing machine didn’t like your offered transaction fee
  • If you don’t pay a sufficient transaction fee, your transaction might float around the network indefinitely
• Few legitimate means of buying or selling BitCoins for other currencies
  • Supposed exchanges accept cash, check, bank transfer, or a medley of unpopular secondary exchanges (no credit cards or PayPal)
  • None of those options is inherently reliable for online purchases or very good for anonymity
  • This really raises the stink level of the whole enterprise

Resources:

• http://www.quora.com/Is-the-cryptocurrency-Bitcoin-a-good-idea/answer/Adam-Cohen-2
• http://falkvinge.net/2011/05/29/why-im-putting-all-my-savings-into-bitcoin/

Basic Stats

• Released: 3/10/2011 on XBLIG @ 80 MSP ($1)
• Dev Tools: XNA 3.1/4.0, Paint.NET, Milkshape 3D, SmartSVN
• Work Effort: 200 hours of tasks, 50+ hours of bug-fixing and polishing
• Sales Expectation: 1000+ downloads, 100+ sales
• Sales Reality: 33000+ downloads, 6000+ sales (in the first week)
• Intent: Develop a novel realistic control scheme, publish a finished indie game
• Download Marksman: Long Range on the Xbox Live Marketplace.
• Some development pictures and screenshots

Good

• Realism in control scheme. Having recently become personally interested in shooting sports and firearms in general, the primary design goal for the game was to demonstrate a more realistic rifle simulation as compared to the “point and click” firing mechanic almost every FPS game uses. A real rifle is a physical artifact which must be held with imperfect fleshy meat-limbs, and the trigger is a mechanically attached component of the entire rifle. The design goal of a physical trigger is to make the gun fire when you want it to and to have the gun not fire when you don’t want it to, thus the trigger has mechanical resistance and pulling the trigger too violently will affect the overall orientation of the entire rifle, enough to make a noticeable difference at a significant distance like 100 meters. Obviously, this would be difficult to achieve with a mouse and keyboard, so I targeted the analog trigger and thumb-sticks of the Xbox 360 controller. The trigger measures analog position, and over time you can calculate velocity and acceleration to examine jerky trigger action and feed into the aiming deflection. Similarly, the game uses a thumbstick to measure “grip stability” – if you adjust your grip while pulling the trigger then the rifle will move. I was hoping to use an analog face button instead of the thumbstick, but the Xbox 360 has only binary face buttons (unlike the original Xbox and PS3). I only continued with the remainder of the game once I was satisfied with the control prototype. The control scheme manages to be challenging to master but fun to use once familiar, just like real firearms!
• Simple & straightforward game concept. Marksman uses a relatively simple central game mechanic: aim and shoot stuff. There is no player location movement, very little game state to manage, and only one bullet is in the air at a time (no full-auto here). This makes it relatively easy to learn, although there is more to a precision rifle shot than just “point and click”. The simulation of the bullet trajectory is quite realistic; it uses real empirical data for the simulated bullets, although it lacks certain minute details such as spin drift (caused by gyroscopic destabilization as a bullet tilts along its trajectory) and the coriolis effect.
• Simple art & sound assets. Almost all the content in the game was produced by me, at my computer desk. This includes all the 3D models, all the textures aside from the skybox, and all the sounds aside from the gunshot (used with permission, and freeware, respectively). Keeping the content simple meant I didn’t have to worry about polishing it too much, or paying to contract it out to someone with more talent. That said, I did perform a polish pass, the ground and backstop were green and gray checkerboards for a long time, and the bullseye and other target textures were re-made with better anti-aliasing and color selections. The bullet model and texture were somewhat painstakingly made to be as close to physically realistic as possible, including the rifling imprint lines which should roughly match the spin rate of the bullet.
• Trial mode. XBLIG trial games have approximately 5 minutes from launch to sell a user on the game; after a certain amount of time, the system pauses the game, cuts off input, and displays a dialog prompting the user to either purchase or quit back to the dashboard. The Marksman trial mode locks out the aspects of the game that might not be fun in a hurry: the highest difficulty, the longer distances, the slow-paced “rifle golf” game mode. The player gets right into the game (optimally picking either time-trial or zombie-horde game modes, although free-play is also available). The longer distances and the harder-hitting high-caliber rifles are also locked out as sort of a teaser. You can highlight the locked rifles in the menu and see their details and stats, but you can’t select or use them.
• Luck. The XBLIG certification process is community-driven, and when the final “pass” vote is given, the game automatically goes live on the marketplace after a short propagation delay. For Marksman, this meant a Thursday release window, ideal for garnering a few sales before the weekend and getting onto the top-selling list when people are at home looking for a new game to play. There were also few or no other competing new releases during that first weekend; combined with a sales data maintenance downtime the subsequent week, Marksman stayed at the top of the best-selling list for a week straight, whether it was actually the top-seller or not. For all the games released during this sales maintenance window, my good luck has been their poor luck; I wish them well, all the same. Luck seems to play a significant role for some XBLIG titles, but the keywords “avatar” and “zombie” also seem correlated with success.

Bad

• Challenge in control scheme. Some people don’t understand how to play the game, since it’s quite different from CoD or Halo, as evidenced in the reviews linked below. The “aim” button is the left shoulder instead of the left trigger (left trigger is “zoom”), so some people flail around for a considerable time unable to even fire a single round. The trigger is treated as analog although most other FPS games present it as a binary button, which potentially adds more confusion. The grip-stability mechanic calls for the right thumbstick to be held still somewhere in a “sweet spot” which is actually donut-shaped to avoid the dead spot at the center and the rim of the stick boundary where the stick can “rest”; this control mechanic is intentionally fatiguing, but might be excessively so.
• Confusing UI. This goes along with the previous point, but is focused on the visual UI rather than the control scheme. Several hints and explanations are present in the game, including a visual “control map”, a multi-page “how to play” description, briefing text at the beginning of each game, and goal text visible at the top of the screen during the entire game; most people completely ignore all of these cues. In long-range shooting, a “drop table” is used to estimate the adjustments needed to place a bullet on-target; the drop table describes the drop of the bullet relative to the point of aim at various distances; real drop-tables are typically produced by a ballistic calculator with some inputs of empirical data for the specific firearm and cartridge being used. The in-game drop table is accessible via the Y button at any point in the game, and displays a pre-calculated set of data for the current rifle/cartridge configuration, and it appears as a large 2-dimensional table of numbers. The other tool available at any time is a “range estimator”, which is in fact a calculator designed to aid in the estimation of target ranges using the mil-dot reticle, but that isn’t adequately described in-game for people unfamiliar with the concept (almost everyone); most people assume this is an automatic laser range-finder, and are confused when it’s wrong. The grip-stability indicator is also not adequately explained, but is rather designed to be plainly visible to someone already familiar with what it is communicating.
• Simple art & sound assets. The menu has no sounds, there is no music in the game, and the scenery on the shooting range is quite sparse (only a ground plane, backstop, target, and skybox). If you listen closely, you can tell the recording quality on some of the sound effects is relatively low, because they were recorded in a computer room with a tile floor and a few blankets lying around. The overall lack of high production values makes it clear this is an “indie game” but doesn’t appear to detract too much from the experience; it may actually keep expectations sufficiently low, along with the price-point. The menu actually consumed a considerable amount of development time, and has some nice background art, so the game clearly has some degree of polish.
• Failed peer-review submission. The game was first submitted for peer-review in late January, in the interest of passing before the XNA 3.1 submission grace period ended on February 7th. It sat in the review queue for two weeks before anyone gave any feedback, and a crash bug was quickly discovered thereafter, which failed the game from review. The specific bug was a failure to handle an exception raised when trying to play too many sounds, and was fixed quickly and easily. Failing from review induces a 1-week wait period before re-submission, which kicked the submission window out past the 3.1 grace period and thus required the game to be ported to XNA 4.0. The process of porting took me a couple of long evenings, and is akin to converting from D3D9 to D3D10 with the similar transition from render-states to state-blocks, but the most painful and time-consuming aspect of the port was the switch to pre-multiplied-alpha by default (my skybox apparently had an all-zero alpha channel which took a few hours to identify). The porting process also exposed a few opportunities for extra polish in rendering, UI layout, and other subtle details. After porting and re-submitting on February 19th, the game again sat in the queue for two weeks before being paid any attention, and eventually passed review on March 10th.

Take-aways

• XNA toolset is fairly robust but also quite basic (D3D + D3DX + content pipeline)
• XBLIG submission process is time-consuming and somewhat subjective
• Xbox dashboard prominence is a vital component of marketing
• I hope this game encourages more people to try shooting real firearms, so long as they handle them safely

Trivia

• The altitude setting, AI AWSM rifle, and 2500m distance were added to replicate the recent world-record sniper shot.
• The ACOG for the M16 is the only non-crosshair scope reticle in the game. The ACOG is correctly calibrated for the simulated M193 cartridge, although the real ones are typically calibrated for M855. The cross-hatches for 400m and beyond are matched to the width of a man’s shoulders (about 19 inches) at the corresponding distance.

Reviews

• User feedback star-rating on XBLIG: ~3.2 (out of 5)
• Video review by youtube user “aaronthesplazer”. No commentary but obviously doesn’t understand control scheme, skips briefing text, ignores goal text at top of screen, quits before trial time-out, gives game a “skip it” rating.
• Video review by youtube user “mainmarco”. Voice commentary demonstrates confusion over control scheme even after briefly viewing the control map, and continual confusion by “inaccurate” bullet-strikes while pulling the trigger. Reviewer skips briefing text, ignores goal text at top of screen in zombie game mode and even exclaims “how do you kill them?” while the text visible in the video says “shoot zombies in the head to kill them”. Gives game a video upload comment of “This game is way to [sic] complex and dumb”, and a verbal rating of “3 out of 10″ or “below average”.
• Written review by “Choke Bunny” of CoedMagazine.com. Excerpt: “This game is for the type of person who claims Call of Duty isn’t realistic. For everyone else, it lacks the bells and whistles to merit even the slightest attention. Well made, but a dull experience overall. BUY or PASS? PASS”
• Written review on “The XNA Roundup”. Generally positive review, the first I’ve seen where the reviewer seems to have actually understood the game design. Excerpt: “You have 8 real life rifles to choose from, each handles differently. There are three difficulty levels, which increase the wind speed, distance and steadying you have to do. The game isn’t flashy, but it wasn’t trying to be, and it works.”

I’ve seen a proliferation of commercials and ads lately for gold-peddling merchants. Some want to buy your unwanted gold jewelry, some want to sell you pure gold coins. They try to pull you in with some simple hook, but the market of precious metals is hardly simple. We are currently experiencing the highest unit price of gold in history – over $1100 per troy ounce – which is good if you bought a bunch of gold 5+ years ago and would prefer to have cash now, but who knows where the price will go from here. This also raises a question as to why the price has skyrocketed recently and why it’s being pushed so hard by the aforementioned merchants.

(more…)

After using my .22 rifle and AR-15 for a while, I got the itch for long-range precision shooting. The .22 cartridge is only accurate to 100 yards, and .223 is accurate to about 600 yards, although my AR-15 isn’t super-accurate. Drawing inspiration from movies and games, I wanted a rifle that could reach out to 1000 yards or more; for a bullet to remain accurate at long range, it must maintain supersonic speed because the transition to subsonic randomly and significantly destabilizes a bullet. Military snipers have used the M21, M24, and M110 to great success, and all of those rifles fire the .308 cartridge which is known to maintain supersonic speed beyond 1000 yards. The M24 is a bolt-action rifle, which is a design well regarded for its accuracy, whereas semi-auto actions like the M21 and M110 are typically regarded as good but not great. I wanted to have an M24 of my own, capable of hitting a dinner plate at 1000 yards.

(more…)

A highlight reel of the last decade of my life, it’ll be hard to top this one!

2000:
Started working on Destiny3D

2001:
Graduated high school
Worked at Fuddruckers
Started university at Georgia Tech

2004:
Internship at Autodesk

2005:
Graduated from GA Tech
Visited Hawaii and England
Started a job at Microsoft, on Direct3D
Started skiing regularly

2007:
Met Sarah

2008:
Visited Tokyo, Japan
Sarah moved in
Got a quaker parrot

2009:
Got into shooting guns, bought several
Started rock climbing
Transferred to Project Natal
Bought a house
Got a dog
Got a cat

Prior to my shotgun acquisition, I was also looking at handguns, as they are far more compact and suitable to carry on the street or in a car. I visited the local indoor range and tried a variety of rentals: 3 different calibers (9mm, .40, and .45), 5 handguns in each caliber, firing 10 rounds per handgun. My experience was that there was a ton of variety in the feel & performance of the available options, so it was tough trying to decide what I liked best. While not exactly scientific, my investigative exercise helped me pare down the selection to examine more thoroughly.

(more…)

After enjoying my two rifles (the .22 and AR-15) for a couple of months, I felt compelled to own a shotgun. The selection of shotguns is quite broad, with different calibers, barrel lengths, and action types, all with somewhat specific purposes in mind. Longer barrels are appropriate for longer-range shooting, and semi-auto actions are good for rapid fire while pump action is meant to be utterly reliable. My intention was to own a shotgun with a primary purpose of home defense, and a secondary purpose of fun at the range.

(more…)

Continuation of series. Previous post.

The .22 rifle is a nice gun, but I wanted something with more reach, more punch, and more reliability – something capable of taking large magazines and spending them at a relatively high rate of fire and with good accuracy. At the time, there was much rumor of the political powers reinstating an assault-weapons ban, and the rumor was strengthened by the attorney general outright proclaiming that goal. That rumor has since faded, but at the time I wanted to hedge that bet by buying something that could be banned, despite the price markup that dealers were enjoying. I simply had to get an AR-15 of reputable manufacture and features.

(more…)

I tried target shooting earlier this year, and discovered how exhilirating it can be to send some hot lead flying down-range. After taking an outing with some friends and trying various pistols and rifles at the local range, I decided I had to buy some of my own that were even better than the ones available for rental. Little did I know how much money I’d end up spending on this hobby in the course of 9 months.

I started with a .22 rifle, the Remington 597 in synthetic/stainless. The actual rifle I received was old stock sold as new, and I discovered it was 10 years old by searching for the serial number. This meant it came with the original-design plastic magazine, which has a tendency to jam up; a quick call to Remington had them sending me a new-design aluminum magazine for free that has been much more reliable. There are also a few rust spots on the barrel and magazine.

The rifle shoots nicely up to about 100 yards and can be taken to indoor pistol ranges without difficulty. I tried using it to complete a marksmanship evaluation, but wasn’t able to do the prone portion of the test due to backstop and target placement. Extrapolating the standing and sitting portions placed me into the top tier of performance, Expert, which made me feel pretty good about both myself and my equipment. My girlfriend also likes shooting this rifle because it has almost no recoil, is relatively quiet, and it has a very simple blowback recoil action. It does jam up once in a while, the most common being a bolt hold-open failure when the magazine is empty, and the occasional failure-to-feed. I wouldn’t trust my life to this rifle, but it’s fine for short-range target practice with cheap ammo.

Decent .22 ammo has been difficult to find since I bought the rifle, and I recently used up the last of the 525-round Federal value pack I bought along with the rifle back in March. I like this particular ammo because it comes in a hefty value pack, it has copper-plated hollow-point bullets, and the price is right around 4 cents per round. Thankfully, Cabela’s had a bulk offering at a decent price with free shipping, so I now have over 4000 rounds of .22 sitting around, begging to be used. I added a cheap Bushnell scope which makes it relatively easy to get nice tight groups at 50-100 yards off a bench rest. The most affordable piece of my arsenal, this .22 rifle has only cost me about $500 in total, including freight, ammo, and cleaning supplies.

Here’s a picture of me shooting the Remington 597 at the local indoor range, Wade’s.